parents' perceptions of standardized testing: its ... · parents' perceptions of...

75

Parents' Perceptions of Standardized Testing: Its RelationshipAnd Effect on Student Achievement

Monica Z. OsburnUniversity of North Carolina at Pembroke

Charles Stegman, Laura D. Suitt, & Gary RitterUniversity of Arkansas

AbstractQuestions regarding the value of standardized testing have been raised bycommunity and school leaders, as well as parents and members of the media.Some have expressed concern that children today are placed under suchpressure to perform well on standardized tests that the anxiety adversely affectsperformance outcomes. This study examined the attitudes and perceptions ofparents with respect to the importance of tests, the responsibility shared byparents and teachers for student performance on these tests, the testing climatesurrounding these tests, and the stress and anxiety that may accompany thesetests. Furthermore, the study investigated the relationship between parentalviews and children's performance on the exams. The sample of studentsexamined here attend school in a high-achieving district in NorthwestArkansas. Parents of fifth-graders, who had just completed the SAT-9 testweek, were surveyed. Overall, parents' responded that standardized testingis important to them and that it is not overly stressful for their children. Theydid, however, report a concern that the teachers are under pressure. Withrespect to student performance on the exams, the parents of students who didpoorly on the exams did feel pressure to help their children do well.

Journal of Educational Research & Policy Studies

76Standardized testing has been placed under significant scrutiny in recent

years. For the better part of the past two decades, schools have implementedlarge scale standardized testing programs as a way to measure academicaccountability (Bernauer & Cress, 1997). In addition, the No Child Left Behindlegislation has increased the importance of standardized testing andaccountability for schools and teachers. In response, administrators, teachers,and parents have questioned the value and worth of tests as a measure ofacademic achievement (Bernauer & Cress, 1997; Paris, 1992). Murray (1998)highlighted the unreasonable emphasis placed on standardized tests by "thegatekeepers of American Higher Education."

In another case for support against testing, Howard Gardner, famous forhis work on multiple intelligences, stated he was unconcerned that Americanchildren were ranked last among the major industrial nations in the ThirdInternational Mathematics and Science Study (Murray, 1998). He reported thattests measure exposure to facts and skills not whether or not kids can think(Murray, 1998).

In many cases, test results are used to make decisions about teacher pay,school funding, and the promotion of children to higher grades (Kubiszyn &Borich, 2000). Given the "high stakes" that are put on these tests, and the vastamount of backlash surrounding standardized testing, it is not unlikely that ahigh level of concern and debate exists. The common concerns highlighted inthe literature are that (a) teachers will teach "to the test" instead of focusing onestablished curricula (Jones, Jones, & Hardin, 1999); (b) students who do notrespond well to standardized tests will be penalized (Etsey, 1997); and that (c)parents, teachers, and administrators can "influence" a child's performancethrough anxiety, pressure, and involvement. These concerns are compoundedfurther when popular press articles, such as Testing Testing Testing, featured inGood Housekeeping (Cool, 2002), highlight standardized testing as stressfulfor parents, teachers, administrators, and students.

Climate and PressureThe concerns raised regarding testing and achievement often are linked

back to not only the validity of the measure, but also its effect on individualsinvolved with testing. Some researchers have examined overall school climate,including teacher anxiety, student pressure to succeed, and administrator's useof testing results, as possible predictors of academic achievement. In a studyconducted by Campbell and Mandel (1990), lower levels of help, pressure, and

Spring 2004 / Volume 4, Number 1

77monitoring, together with higher levels of psychological support from parentswas associated with higher academic achievement. Pang's (1991) study includeda variable of parental support as part of climate and found that, when studentsperceived their parents as supportive, mathematics achievement increased. Theliterature, however, also supported the claim that testing environments werehigh anxiety and extremely stressful for students.

In North Carolina, one survey found that 61% of teachers perceived thattheir students felt more anxiety and less confidence due to testing (Jones et al.,1999). Some researchers have found that test anxiety starts as early askindergarten and continues throughout the testing process (Fleege, Charlesworth,Burts, & Hart, 1992; Hill & Wingfield, 1984). This anxiety is often compoundedwith successive low-performing testing results (Crocker, Schmitt, & Tang, 1988).As anxiety increases, students may look to parents for increased support.

A recent survey sponsored by the Association for Supervision andCurriculum Development found that many parents are confused aboutstandardized testing, do not feel informed about assessment procedures and donot believe they are equipped to assist their child in preparing for testing(Gleason, 2000). Dounay (2000) stated that parents in some states assert thathigh-stakes tests place undue pressure on young children and these parentshave questioned the validity of assessment and accountability systems. Nowherein the current literature was parental stress actually measured.

Parental InvolvementResearch has indicated that parental involvement can be a significant

factor in predicting academic success (Christenson, Rounds, & Gorney, 1992;Epstein, 1987; Keith, Troutman, Bickley, Trivette, & Singh, 1993). However,the construct of parental involvement has been measured in several differentways. Some early studies such as Morrison (1978), conceived parentalinvolvement as the extent to which parents were involved in school activities.

More recent studies, however, have expanded the scope of the construct.Milne, Myers, Rosenthal, and Ginsburg (1986) used measures of whether parentshelped with homework or attended parent-teacher conferences as variables tomeasure parental involvement. Fehrmann, Keith, and Reimers (1987)conceptualized a measure of parental involvement that included whether parentsknew where their children were and what they were doing, whether parentsinfluenced their children's plans after high school, and whether parents closelymonitored how well their children were achieving in school.


78Astone and McLanahan's (1991) study included a measure of general

supervision and measures of whether the parents had high aspirations for theirchildren, monitored school progress, and talked regularly with their children.Miller and Green (1992) used a structural equation model with parentalinvolvement represented by measures of whether parents helped withhomework, rewarded good grades, and communicated with teachers aboutschool matters. Madigan (1994) examined the effects of 10 indicators ofparental involvement associated with whether parents encourage and rewardwork on mathematics at home. Most of the previous studies employedmultivariate regression analysis and estimated effects over and above familybackground factors such as socioeconomic status.

Not all studies on parental involvement indicated relationships that werepositive. Milne et al. (1986) found a negative relationship betweenachievement and parental help with homework and suggested that this wasattributable to the fact that parents helped more if their children were notdoing well at school. Madigan (1994) found negative associations betweenparents helping with homework, insisting that their children do theirhomework, or rewarding their children for good grades and academic success.Nonetheless, both Milne et al. (1986) and Madigan (1994) found a positiverelationship between parents' high expectations for their children and academicperformance. Muller (1993) also reported negative relationships toachievement with parents monitoring their eighth graders' homework orproviding more after school supervision. Significant negative effects forparents' frequent contact with school or participation in parent-teacherconferences were also found. Muller (1993) reported a link between parentsdiscussing school experiences with their children and increased schoolperformance. Astone and McLanahan (1991) found a positive correlationbetween parents' aspirations, monitoring their child's progress, and generalsupervision, with academic achievement. Fehrmann et al. (1987) foundpositive relationships with his parent involvement construct consisting ofmonitoring and supervision.

In summary, studies found that parents' high expectations for and generalmonitoring of their children's performance were positively related to academicachievement, whereas helping with homework and attending parent-teacherconferences have negative associations or no association with academicachievement. What this review suggests is that parent involvement is a multi-dimensional construct that can have many components. Some types of


79involvement may well lend to better outcomes, while other types of involvementmay well arise due to poor performance. Rarely in the literature was parent'sperceptions of test value evaluated nor was the test climate factored into theresearch.

Test ValueLittle research to date has studied parents' perceptions of achievement testing.

Barber, Paris, Evans and Gadsden (1992) indicated that parents supportachievement testing. However, too often parents are not informed of results,the results are not explained adequately, and the results are often difficult tointerpret without assistance. Paris (1992) found that most states do not haveformal policies for communicating test results to parents, and when surveyed,many teachers and administrators admitted being unable to interpret results.

If parents do not understand achievement testing, this may underminetheir support. It is unclear if parents understand academic testing and it is alsounknown whether this lack of understanding affects student performance. Aparent may be involved, but if that involvement is negative toward achievementtests or promotes an unhealthy environment for test taking, how might thatimpact academic performance? The literature highlights types of parentalinvolvement and perceptions of a stressful academic environment, but it hasnot systematically examined parents' perceptions of standardized testing as itrelates to student achievement.

This study will help address questions related to parental perceptions oftest value, their perceived role in testing, and how that is related to students'academic achievement. It is hypothesized that parental attitudes towardstandardized tests and test climate are related to student academic performance.According to Astone and McLanahan (1991), a child's failure in school is partlythe result of inadequate or ineffective parenting styles. Specifically, a childmay be influenced through the parent's perceptions, and thus may or may notsucceed academically. Astone and McLanahan (1991) found that several parentalpractices (such as parents' aspirations toward school, monitoring academicprogress, general supervision, and talking with children) had significant effectson student grades, attitude toward school, and graduation rate. This paper seeksto add the factor of climate to the variables studied by Astone and McLanahanas well as focus on standardized assessments as a dependent variable. Thequestions guiding this study are the following: (a) Do parents believe thatstandardized testing is valuable and are parents interested in the performance


80of their children on standardized tests ?; (b) Is there an "unhealthy" school climatedue to testing ?; (c) Are parents concerned that standardized testing is overlystressful for their children or for the teachers of their children and do they feelpressure related to the performance of their children ?; and (d) Is there a relationshipbetween student performance on standardized tests and any of the above attitudesand perceptions of the parents?

Methods

ParticipantsParents of children attending fifth-grade classes in a high-performing

school district of Northwest Arkansas were selected to participate. Thisparticular district ranked in the top 10% of all the districts in the state on mostachievement tests. Further, the Northwest region of Arkansas is one of themost affluent in the state. The district has nine schools that serve fifth grade.Each school was similar with respect to diversity (predominately Caucasian).The individual schools exhibited some variability in socio-economic statusand family structure. All parents were provided with a packet that containedan explanation of the study, a consent form, and a parent survey. A total ofapproximately 500 parents received surveys and 250 parents returnedcompleted surveys. Of the 250 surveys returned, 190 could be matched withstudent achievement data.1 These 190 surveys were from 9 different schoolsand 23 total classrooms. The drop off in matching was due to errors in testreporting and incomplete testing data, and there is no reason to believe thatthe respondents are systematically different from the non-respondents.

MeasuresData for this analysis were collected by means of a self-report survey

completed by the parents and the student Stanford Achievement Test, Version9 (SAT-9) scores. The SAT-9 tests were taken the same week the parent surveyswere distributed and returned. The authors designed the self-report survey,Parent's Survey of Standardized Achievement Tests (see Appendix A). Thesurvey consisted 19 items relating to perceptions of standardized testing,

1With a sample of 190, the authors will be able to detect, with statistical significance,a difference of approximately .4 effect sizes (using a 95 percent confidence level and 80percent power). For example, on the Math portion of the SAT-9 exam, this design willallow for us to detect differences of approximately 15 scale points.


81specifically, overall stress and anxiety, climate, and parental involvement. A five-point Likert-type scale was used to rate responses ranging from not at all (1) toan extreme amount (5). Survey items were categorized into three scales orconstructs: Parental Involvement and Interest in Testing, Testing Climate, andOverall Stress and Anxiety.

The Parental Involvement and Interest in Testing construct included thefollowing items: (a) The standardized testing program is important for theeducational progress of my child; (b) I believe that standardized testing is awaste of time (This question is reverse coded to fit in with the construct); (c)I am interested in the results of my child's tests; and (d) I believe that parentshave a responsibility to work with their children to improve their performanceon standardized tests.

The Testing Climate construct included the following items: (a) Theclimate surrounding testing in this school is healthy; (b) I think the teachersgenuinely want my child to do well on the test; (c)The principal works hardto help make the testing week as pleasant as possible for the students; and (d)The principal works hard to help make the testing week a positive experiencefor the students.

The Overall Stress and Anxiety construct items were: (a) The standardizedtesting program is stressful for my child; (b) The standardized testing programis stressful for teachers; (c) Teachers seem threatened by the testing program;and (d) I feel pressure to help my child score well on standardized tests.

For the construct of Parental Involvement and Interest in Testing, theitem reliability was .70; the Testing Climate construct yielded a .79 itemreliability; and the Overall Stress and Anxiety construct yielded a .71 itemreliability. For each construct a mean score resulted between one and five, inwhich a score near one suggests a low score and a score near five means theyscored high on that construct (for example, on the Testing climate construct,a score of one means that the parent felt the climate was not positive at alland a score of five means they felt the climate was extremely positive). Tomake sense of the construct scores, the authors grouped them into threecategories such that the lowest grouping included scores nearest to one; themiddle grouping included scores closest to two and three; and the highestgrouping included scores nearest to four and five. The mean score for eachconstruct was then used to classify each respondent as a low scorer (0 to 1.49= did not agree at all), middle scorer (1.50 to 3.49 = some or moderate


82agreement), or high scorer (3.5 to 5 = significant or extreme agreement) on thatparticular construct. The authors chose to categorize the scores into three groupsinstead of five so they would be easier to understand. This in no way changedthe results. For example, if the results were significant on the 1 to 5 scale theywere also significant when the scores were categorized and vice versa. TheStanford Achievement Test version 9 (SAT-9) was part of the standardized testingprogram required by the Arkansas Department of Education when this studywas completed. The SAT-9 battery used consisted of three content areas,mathematics, reading, and language.

ProceduresElementary school principals in the district were approached at the

beginning of the academic year of 2000-2001 and approval was given toadminister the survey to students, teachers, and parents in this district. Theweek prior to the administration of the SAT-9, all fifth-grade teachers wereinstructed to send home a permission slip for children to participate in a studyon standardized tests. Attached to that permission slip was an informed consentas well as the Parent's Survey of Standardized Achievement Tests for parentsto complete for participation in the study. Either parent could complete thesurvey, but the one who actually responded, needed to give informed consentand sign the permission slip for themselves and their child. The completedsurveys, informed consent, and permission slips were then returned to thefifth grade teachers who then gave them to one of the researchers. All parentswere given the same amount of time to complete the survey.

The teachers, counselors, and administrators of the school districtadministered the SAT-9 in the Fall of 2000. Each student's SAT-9 scores werematched to the parent's questionnaire by a unique number identifier. A furtherdata check was also conducted by name of teacher and grade of student. Allidentifying data must have matched exactly to have been used in the study.

To address the research questions, several different procedures were used.First, the authors examined the item response by percentage of respondentsto accurately address the scope of the problem. Next, to assess the relationshipbetween the items and constructs and student achievement, a series of analysesof variance (ANOVA) were conducted.


83Results

Parental Involvement and Interest in TestingThe vast majority of parents reported to be involved in and interested in

their child's testing with a mean construct score of 3.94 on a five point scale(5 = extremely involved and 1 = not involved at all). Now using the categoriesdescribed in the methods section, 83% of parents reported significant or extremeinvolvement and interest in their child's testing (see Figure 1).

The individual items do suggest that the parents surveyed reported thatstandardized testing is important. Fewer than 5% reported that it was not atall important, while 55% responded that it is significantly or extremely important.For a related survey question "I believe standardized testing is a waste of time,"two out of three parents felt that testing was not a waste of time.

An overwhelming majority of parents surveyed were interested in the resultsof their child's test scores. Out of the 190 parent responses, 88% reported beingsignificantly or extremely interested in their child's results. Only three parentsreported not being interested at all in the results of the test. Parents surveyedalso felt a responsibility to help their children. In response to the survey item "Ibelieve that parents have a responsibility to work with their children to improvetheir performance on standardized tests," approximately 50% of the parents

Figure 1. How Involved and Interested are the Parents? These figures are based onthe four items from the Parental Involvement and Interest in Testing construct.

116

83

0

2 0

4 0

6 0

8 0

1 0 0

Not a t a l linvo lved and

in te res ted

S o m e w h a t o rmodera te lyinvolved and

in te res ted

S ign i f i can t l y o re x t r e m e l y

invovled andin te res ted

Pe

rce

nt


84agreed strongly (significantly or extremely). In fact, fewer than 7% of the parentsreported to have no responsibility for the child's performance.

Although these items were not included in the formal construct, there weresome indications of negative outcomes related to parents and the testing process.One concern lies with the lack of communication between parents and schoolstaff regarding test results. Supporting the findings of Barber et al. (1992), halfof the parents reported that they had received only some or no explanation of thetest score results by their child's teacher. One possibility was that the schoolcounselor had undertaken the task of explaining these results. More than 90%of the parents, however, reported that they have received only some or noexplanation of the test score results by the school counselor.

Testing ClimateMost parents reported the testing climate as being relatively positive (M =

3.84 on a five point scale, 5 = extremely positive and 1 = not positive at all). Infact, based on the construct, 75% of parents reported that they felt their child'stesting environment was significantly or extremely positive (see Figure 2).

With respect to the individual items, a majority (59%) of the parents reportedthat the climate surrounding testing was healthy. Fewer than 4% of the parentsreported that the testing climate was not at all healthy for their children.Likewise, approximately 86% of parents think that teachers want their child to

1

23

75

0

20

40

60

80

100

Not at a l lpositive

Somewhat o rmoderately

positive

Signif icantly orex t remely

positive

Pe

rce

nt

Figure 2. How Positive Is the Testing Climate? These figures are based on the fouritems from the Testing Climate construct. Two percent of the responses for this constructwere reported as missing data.


85do well on standardized tests. Moreover, the majority of parents (64%) reportedthat the principal of the school (in each of nine different schools) works hard tomake the testing week a pleasant and positive experience for the students.

Overall Stress and AnxietyMost parents reported that stress and anxiety was somewhat or moderately

present (M = 2.41 on a five point scale, 5 = extremely stressful and 1 = notstressful at all). Approximately 73% of parents reported the stress and anxietylevel to be moderate (see Figure 3).

Regarding the individual items, the majority of parents expressed the beliefthat the testing program at the school is not overly stressful for their children.In response to the survey item "The standardized testing program is stressful formy child," only 24% responded that it is significantly or extremely stressful. Infact, 22% of the parents reported that it was not at all stressful for their children.With respect to the teachers, a roughly similar proportion of the parents believedthat teachers are concerned about the pressure due to the standardized tests.

Just as most of the parents did not believe that the testing program is overlystressful for their children, the majority of parents reported that they did not feelpressure to help their children score well on the exams. In respect to the surveyitem "I feel pressure to help my child score well on standardized tests," only13% reported significant or extreme pressure. In fact, 43% of the parents claimedto have felt no pressure at all.

1 0

7 3

1 7

0

2 0

4 0

6 0

8 0

1 0 0

N o s t r e s s a n da n x i e t y a t a l l

S o m e w h a t o rm o d e r a t e l e v e lo f s t r e s s a n d

a n x i e t y

S i g n i f i c a n t o re x t r e m e l e v e l o f

s t r e s s a n da n x i e t y

Pe

rce

nt

Figure 3. What is the Overall Stress and Anxiety Level? These figures are based onthe four items from the Overall Stress and Anxiety Construct.


86Relationship Between Parental Attitudes and Student Achievement

Not only did this study seek to understand the attitudes of parents towardstandardized testing, but also to examine the relationship between parentalattitudes and student performance on standardized tests. Relationships betweeneach of the construct scores and student achievement were examined. Severalspecific items for relationships between parent responses and student achievementscores were also examined. An alpha level of 0.05 was used for all univariateanalyses.

Results for the Parental Involvement construct indicated no significantdifference in student achievement based on mathematics [F(2,187) = 1.00, p <0.37] or reading [F(2,187) = 1.98, p < 0.14], but there was a significantdifference for language [F(2,183) = 3.23, p < 0.04]. Students of parents whowere not interested or involved at all scored the highest on language with amean score of 665 and parents who were moderately involved and interestedhad students who scored the lowest in language with a mean score of 636 (inthe state of Arkansas the range of mean scores for language was 508 to 773, seeAppendix D for more information). For the parents who were significantly orextremely involved and interested, their students had scores that fell in the middlerange for the language section of the test with a mean score of 653.

There were no significant differences found on the Testing Climateconstruct for student achievement based on mathematics [F(2,183) = 1.11, p <0.33], reading [F(2,183) = 0.87, p < 0.42] or language [F(2,179) = 0.39, p <0.68]. A significant difference was reported for student achievement based onmathematics [F(2,187) = 3.66, p < 0.03] on the Overall Stress and Anxietyconstruct. There were no significant differences for student achievement basedon reading [F(2,187) = 1.40, p < 0.25] and language [F(2,183) = 2.03, p <0.13] for the Overall Stress and Anxiety construct. Students of parents whoreported no stress or anxiety scored in the middle for mathematics with a meanscore of 651 (in the state of Arkansas the range of mean scores for mathematicswas 494 to 784, see Appendix D for more information). Parents who reportedmoderate levels of stress and anxiety had students who scored the highest inmathematics with a mean score of 658. For the parents who reported significantor extreme levels of stress and anxiety, their students had the lowest scores witha mean score of 636 on the mathematics section of the test.

While there were few significant parent results revealed in the constructabove, some of the individual items were related to student achievement ininteresting ways. For the item "I feel pressure to help my child score well," one-


87way univariate ANOVA's indicated that there were significant differences betweenthe parent groups based on responses to the amount of pressure and studentachievement in mathematics (F(2, 186) = 7.39, p < 0.00), reading (F(2,186) =9.27, p < 0.00), and language (F(2,182) = 9.78, p < 0.00). One might havehypothesized that parents of students who performed poorly would feel a higherlevel of pressure, or that students with more "interested" parents would haveperformed better on exams. Data gathered suggested that parents who reportedfeeling pressure to help their children perform well on exams had children withthe lowest average scaled scores (see Table 1).

Table 1. Relationship between Parents' Pressure and StudentPerformance

Parental Reports of Math Reading Language“I feel pressure to help Scaled Scaled Scaledmy child score well” Score (M) Score (M) Score (M)

Pressure: Not at All (n=78) 660 675 656

Pressure: Some or Moderate (n=88) 655 667 652

Pressure: Significant or Extreme (n=23) 624 634 622

Overall Group Averages 653 666 650

ANOVA F-statistic Significance .001 .000 .000

As Table 1 indicates, the students whose parents reported no pressure toimprove student performance represented the highest scoring group on all threeexams. For example, on the mathematics exam, children whose parents reportedno pressure had an average score of 660, while students whose parents reported"some or moderate" pressure had an average score of 655 and the students ofthose parents who reported "significant or extreme" pressure had a much loweraverage scaled score of 624. This same pattern was evident for the reading andlanguage portions of the SAT-9 exam as well.

No significant differences were found for the item "The standardized testing

Note: Authors' analysis of parental survey data and SAT-9 test scores.


88

Discussion

Academic testing is currently controversial in educational systems. Theresults of this study provide some additional information that should be useful.Despite what the popular press may imply, this study indicated that parents arenot overly anxious about standardized testing, nor do they feel their childrenare. However, the district used in this study is a fairly high-performing district,with only a couple of low-performing schools. Parents from districts withprimarily low-achieving schools may respond differently.

One interpretation of this interesting result is that parents of the lowest-performing students react in one of two ways: (1) they reveal a belief that thesetests are not meaningful, perhaps because they do not think their children dovery well, or, (2) they emphasize the importance of the exam due to the greatconcern they may have over the performance of their children. These results,however, should be viewed with caution as there is no hint of causation ineither direction; that is, there is no evidence as to whether the student performanceinfluenced parental views or vice versa.

program is stressful for my child" between the parent groups based on responsesto the amount of pressure and student achievement in student achievement inmathematics [F(4,184) = 2.33, p < 0.06], reading [F(4,184) = 1.26, p < 0.29],and language [F(4,180) = 2.19, p < 0.07]. Students who scored the lowest onthe mathematics, reading, and language exams also had parents who reportedthat the tests were relatively stressful for their students. Whereas the studentswhose parents reported no stress for their children earned an average mathematicsscaled score of 663, their counterparts, whose parents reported "significant orextreme" stress for their children, had an average score of only 635. Thisdifference was somewhat modest and not statistically significant; however, itwas still a difference worth noting.

For the item "The standardized testing program is important for theeducational progress of my child," there was a difference between the parentgroups based on responses to the amount of pressure and student achievementin mathematics [F(4,185) = 3.30, p < 0.01], but not in relation to reading[F(4,185) = 2.15, p < 0.08] or language [F(4,181) = 2.34, p < 0.06]. Aninteresting pattern appeared in which the students whose parents expressed verystrong views - on either the positive or negative side - about the importance ofstandardized testing were among the lowest performers on exams. Further, theparents of students who perform well did not report strong views on the supportfor standardized testing (see Table 2).


89Table 2. Relationship between Parent Test Value and StudentPerformanceParental Reports of"The standardized testingprogram is important for Math Reading Languagethe educational progress Scaled Scaled Scaledof my child" Score (M) Score (M) Score (M)

Pressure: Not at All (n=9) 637 668 648Pressure: Some (n=23) 642 662 640Pressure: Moderate Amount (n=54) 665 674 657Pressure: Significant Amount (n=80) 656 669 652Pressure: Extreme Amount (n=24) 635 645 635 Overall Group Averages (190) 653 666 650

ANOVA F-statistic Significance .012 .076 .057

Note: Author's analysis of parental survey data and SAT-9 test scores.

With respect to the value of testing, these parents indicated that testing isimportant but that they are not getting the results explained to them. To makeachievement testing more valued and beneficial to all involved, it should be acollaborative process between the school and the family. The results indicated apositive perception of school climate, which should support bettercommunication to assist the student academically. The parents also indicated asense of responsibility to help their child succeed academically. This can be seenas a willingness to become more involved in the testing process. This districtmay want to investigate how parent involvement can be beneficial in the testingprocess.

Overall, this study does not support the concept that parents are overlystressed about their children and standardized testing. To the contrary, it showedsupport for standardized testing and reaffirmed its value in the academic process.Future research needs to be conducted to assess if these results would be the


90same in a district that is experiencing significantly lower testing scores andmore pressure to improve current test results. One of the limitations to thestudy was that it only covered one specific region with a small degree of culturalor socioeconomic variation. The results may not reflect a wider population. Forexample, schools labeled "academically distressed" in Arkansas would be goodsites for similar research in the future. Students in Northwest Arkansas generallyscore higher in academic performance than other parts of the state, andspecifically this school district is under little state pressure to increase test scores.A broader study would examine what if any differences are apparent in otherregions of the state.

Another limitation of the study was some of the unknown factors associatedwith parent involvement and perceptions. It is a highly interpretable conceptthat may be influenced by factors such as parental age, education level or familystatus. These variables were not accounted for in the study. The results presentedhere can be used as a springboard for future testing. Moderate levels of parentalinvolvement coupled with parental interest and knowledge may assist in furtheringthe benefits of standardized testing.

References

Astone, N. M., & McLanahan, S. S. (1991). Family structure, parental prac-tices and high school completion. American Sociological Review, 56, 309-320.

Barber, B. L., Paris, S. G., Evans, M., & Gadsden, V. (1992). Policies forreporting test results to parents. Educational Measurement: Issues andPractice, 11 ,15-20.

Bernauer, J. A., & Cress, K. (1997). How school communities can helpredefine accountability assessment. Phi Delta Kappan, 1, 71-75.

Campbell, J. R., & Mandel, F. (1990). Connecting math achievement toparental influence. Contemporary Educational Psychology, 15, 64-74.

Christenson, S. L., Rounds, T., & Gorney, D. (1992). Family factors andstudent achievement: An avenue to increase students' success. School Psychology Quarterly, 7, 178-206.

Cool, L. C. (2002). Testing testing…testing…: Are standardized exams thestrong medicine our schools need? Or will they turn kids off to learning?The bottom line on a raging debate. Good Housekeeping, August.

Crocker, L., Schmitt, A., & Tang, L. (1988). Test anxiety and standardizedachievement test performance in the middle school years. Measurement andEvaluation in Counseling and Development, 20(4), 149-157.


91Dounay, J. (2000). High-stakes testing is high-stress too. The Education

Digest, 65(9), 9-13.Epstein, J. L. (1987). Effects on student achievement of teachers' practices of

parental involvement. In S. Silver (Ed.), Literacy through family,community, and school interaction. Greenwich, CT: JAI Press.

Etsey, Y. K., (1997, March). Teachers' and school administrators' perspectives and use of standardized achievement tests: A review of publishedresearch. Paper presented at the meeting of the American EducationalResearch Association, Chicago, IL.

Fehrmann, P. G., Keith, T. Z., & Reimers, T. M. (1987). Home influence onschool learning: Direct and indirect effects of parental involvement on highschool grades. Journal of Education Research, 806, 330-337.

Fleege, P. O., Charlesworth, R., Burts, D. C., & Hart, C. H. 1992. Stressbegins in Kindergarten: A look at behavior during standardized testing. Journal of Research in Childhood Education, 7, 20-26.

Gleason, B. (2000). National survey gauges parent perceptions of statemandated standardized tests. Association for Supervision and CurriculumDevelopment: Alexandria, VA.

Hill, K., & Wingfield, A. (1984). Test anxiety: A major educational problemand what can be done about it. The Elementary School Journal, 85(1),105-126.

Jones, M. G., Jones, B., & Hardin, B. (1999). The impact of high-stakestesting on teachers and students in North Carolina. Phi Delta Kappan, 81,199-203.

Keith, T. Z., Troutman, G. C., Bickley, P. G., Trivette, P.S., & Singh, K. (1993).Does parental involvement affect eighth grade student achievement?Structural analysis of national data. School Psychology Review, 22,474-496.

Kubiszyn, T., & Borich, G. (2000). Educational testing and measurement:Classroom application and practice (6th ed.). New York: John Wiley.

Madigan, J. (1994). Parent involvement and school achievement. Paperpresented at the meeting of the American Educational Research Association,New Orleans, LA.

Miller, J. D., & Green, H. (1992). The impact of parental and home re-sources on student achievement and career choice. Paper presented at themeeting of the American Association for the Advancement of Science,Chicago, IL.


92Milne, A. M., Myers, D. E., Rosenthal, A. S., & Ginsburg, A. (1986). Single

parents, working mothers, and the educational achievement of schoolchildren. Sociology of Education, 59, 125-139.

Morrison, G. S. (1978). Parent involvement in the home, school, andcommunity. Columbus, OH: Charles E. Merrill.

Muller, C. (1993). Parent involvement and academic achievement: An analysisof family resources available to the child. In B. Schneider & J.S. Coleman(Eds.), Parents, their children, and schools (pp. 77-114). Boulder, CO:Westview.

Murray, D. W. (1998). The war against testing. Retrieved on March 10, 2004from http://www.stats.org/record.jsp?type=oped&ID=85

Pang, V. O. (1991). The relationship of test anxiety and math achievement toparental values in Asian-American and European-American middle schoolstudents. Journal of Research and Development in Education, 24(4), 1-9.

Paris, S. G. (1992). Four perspectives on educational assessment. InternationalJournal of Disability, Development, and Education, 39(2), 95-105.

Appendix A

(Appendix A continiues)

Not at all SomeModerate amount

Significant amount

Extreme amount Total

9 23 54 80 24

4.70% 12.10% 28.40% 42.10% 12.60%

41 55 48 28 17

21.70% 29.10% 25.40% 14.80% 9.00%

14 56 47 41 13

8.20% 32.70% 27.50% 24.00% 7.60%

59 49 36 20 4

35.10% 29.20% 21.40% 11.90% 2.40%

78 43 35 19 4

43.60% 24.00% 19.60% 10.60% 2.20%

54 38 34 45 14

29.20% 20.50% 18.40% 24.30% 7.60%

Parent’s Survey of Standardized Achievement Tests

1

The Standardized testing program is important for the educational progress of my child.

190

2The Standardized testing program is stressful for my child.

189

3The standardized testing program is stressful for teachers.

171

6I have had the results of my child's test explained to me by a teacher.

185

4Teachers seem threatened by the testing program.

168

5I feel pressure to help my child score well on standardized tests.

179


93Appendix A (continued)

(Appendix A continiues)

164 8 2 5 7

88.20% 4.30% 1.10% 2.70% 3.80%

7 18 49 76 31

3.90% 9.90% 27.10% 42.00% 17.10%

3 3 17 44 123

1.60% 1.60% 8.90% 23.20% 64.70%

126 29 21 5 6

67.40% 15.50% 11.20% 2.70% 3.20%

1 6 18 67 93

0.50% 3.20% 9.70% 36.20% 50.30%

39 44 45 30 18

22.20% 25.00% 25.60% 17.00% 10.20%

33 51 38 32 21

18.90% 29.10% 21.70% 18.30% 12.00%

14 48 49 58 20

7.40% 25.40% 25.90% 30.70% 10.60%

7

I have had the results of my child's test explained to me by a counselor.

186

8The climate surrounding testing in this school is healthy.

181

9I am interested in the results of my child's tests.

190

10I believe that standardized testing is a waste of time.

187

11

I think the teachers genuinely want my child to do well on the test.

185

12

I think the teachers are concerned about the test results impacting their job security.

176

13

I think teachers are concerned about the pressure that could be placed on them by the principal if their classes' test

175

14

I believe the teacher is responsible for working with my child to improve his/her performance on standardized tests.

189


94Appendix A (continued)

13 29 53 52 41

6.90% 15.40% 28.20% 27.70% 21.80%

50 41 52 23 20

26.90% 22.00% 28.00% 12.40% 10.80%

0 4 23 85 77

0.00% 2.10% 12.20% 45.00% 40.70%

4 18 48 63 43

2.30% 10.20% 27.30% 35.80% 24.40%

4 18 41 70 42

2.30% 10.30% 23.40% 40.00% 24.00%

15

I believe that parents have a responsibility to work with their children to improve their performance on standardized tests.

188

16

My child likes the testing week because he/she has less homework and less instruction in the class.

186

19

The principal works hard to help make the testing week a positive experience for the students.

175

17My child tries to do well on the tests. 189

18

The principal works hard to help make the testing week as pleasant as possible for the students.

176

Appendix BTable 3. Descriptives of the Testing Variables

Variable Name N M SD Min. Max.

SAT-9 Reading scaled score 190 666.37 42.02 546 776

SAT-9 Mathematics scaled score 190 653.52 41.81 508 762

SAT-9 Language scaled score 190 650 34.59 548 750


95

Appendix C

Table 4. Descriptives of the Constructs

Variable Name N M SD Min. Max.

Parental Involvement Construct 190 3.36 0.70 1.00 5.00Testing Climate Construct 186 3.84 0.78 1.00 5.00Overall Stress and Anxiety Construct 190 2.41 0.88 1.00 4.75

Appendix D

Table 5. Descriptives of the Testing Variables for the state of Arkansas andNationally

Variable Name M SD Min. Max.

Arkansas SAT-9 Reading scaled score 643.54 43.89 488 799 SAT-9 Mathematics scaled score 629.09 39.53 494 784 SAT-9 Language scaled score 631.37 38.30 508 773

Nationally SAT-9 Reading scaled score 648.5 43.8 SAT-9 Mathematics scaled score 635.7 37.8

SAT-9 Languagescaled score 637.6 37.3

parents' perceptions of standardized testing: its ... · parents' perceptions of...

Documents